Platform Briefs

Data Master Extension Overview

Issue link:

Contents of this Issue


Page 0 of 1

ZALONI DATA MASTER EXTENSION Crea ng "golden" or master records from mul ple data sources helps organiza ons achieve a single version of truth, as well as an enriched view of customer or product data for applica ons such as intelligent pricing, personalized marke ng, smart alerts, customized recommenda ons, and more. Zaloni Data Master Extension: • Spark Machine Learning based approach for matching and linking records based on supervised learning techniques • Matches all types of data • Provides an IT and Business interface • Leverages live data to train the matching models • Leverages Spark architecture to scale based on data and cluster size Data Master Extension Overview 1 Critical Automation Without Significant New Investment Tradi onally master data management (MDM) solu ons have delivered golden records with mixed results -- they can be complex, inflexible and deliver low ROI. In contrast, Zaloni Data Master Extension, which is offered as an extension to the Zaloni Data Lake Management Pla orm, enables a prac cal, unique solu on for Customer or Product 360° ini a ves at the scale of big data. Zaloni Data Master Extension leverages the data lake to implement its machine- learning data matching engine, automa ng the capture and combina on of any data type, including unstructured data. This is a new approach in the market for crea ng an integrated, consistent view of data that is efficiently maintained and can drive customer-facing applica ons. THE DATA LAKE COMPANY Powerful Machine-Learning Zaloni Data Master Extension is built on top of the robust Zaloni Data Management Pla orm and uses Spark machine-learning libraries and analy c approaches to integrate data silos, even in the absence of unique iden fiers from opera onal systems. These approaches include probabilis c matching for record linkage, and advanced data clustering and data classifica on techniques. In addi on, the Zaloni Data Master Extension uses reinforced learning techniques that enable customers to train the matching models based on live sample data. This "training" provides maximum accuracy that may be adjusted as data changes over me. If a current MDM solu on is in place, the Data Master Extension can be used to augment the current solu on allowing organiza ons to train the ML algorithms, used to iden fy duplicate or near duplicate data, based on their own live data samples.

Articles in this issue

view archives of Platform Briefs - Data Master Extension Overview